Melanocytic nevi, including dysplastic or atypical nevi (DN), can recur or persist following shave removal procedures, and recurrence may resemble melanoma, both clinically and histologically (pseudomelanoma). Recurrence may originate from proliferation of the remaining neoplastic melanocytes following incomplete removal. The present study determines the rate and etiology of this event. A cross-sectional analysis of 110 excision specimens showing histological recurrence was performed, and these specimens were compared to the slides of the original shave specimens showing mildly atypical DN. In the second portion of the study, a retrospective review of 167 cases with biopsy-proven mildly atypical DN which were followed up for at least two years was conducted to determine the rate of recurrence/persistence. When followed up for two years, DN, with positive shave margins, defined by extension or very close extension (≤0.2 mm) of the lesions to the lateral margins and into the deep margins through the hair follicles in the shave removal specimens, have a higher probability of recurrence than DN with negative (or clear) margins (odds ratio (OR) = 158; 95% confidence interval (CI) = 36.62–683;
Dysplastic or atypical nevi (DN) are one of the most frequently encountered lesions in dermatopathology. The prevalence of histological DN varies by race and ethnicity and can reach up to 50% in some white populations [
Melanocytic nevi (including DN) can recur or persist following shave procedures, and recurrence may resemble melanoma, both clinically and histologically (pseudomelanoma). The majority of recurrences occur within less than six months of the primary shave procedure and only rarely occur after 24 months [
Until April 2015, in DermOne clinics (the location where this study was performed in Forked River, New Jersey), DN with moderate atypia or severe atypia on the shave removal specimens (SRS) were routinely excised regardless of the status of the margins, without providing sufficient time for recurrence. Patients with mildly atypical DN were followed up for up to two years, and the lesions were excised only if repigmentation occurred within or around the scars. However, since April 2015, DN with moderate atypia are excised only if the margins of the shave specimens are positive, and those with clear margins are only excised if they recur. The management guidelines for DN with mild atypia and severe atypia have remained the same.
Goodson et al. found a low (3-4%) rate of clinical recurrence for DN with mild atypia or moderate atypia and benign nevi when following up patients for two years after biopsy [
Although providers frequently ask pathologists to evaluate the margins in melanocytic neoplasms, a survey of more than 150 dermatopathologists revealed that only about one-third routinely comment on the margins of shave or punch specimens. This is probably due to the lack of guidelines for reporting margins of melanocytic neoplasms other than melanomas. Most pathologists, however, comment on the margins of the excision specimens (ES) [
In this study, the author compared the ES that showed evidence of recurrent or persistent DN (which were performed for repigmenting mildly atypical DN diagnosed on the SRS) with the slides of the original shave specimens to determine whether involvement of the margins is associated with recurrence.
A cross-sectional analysis of 110 ES of biopsy-proven mildly atypical DN from 87 patients in more than 20 clinical offices throughout the state of New Jersey between September 2014 and December 2014 was performed. The selected ES were elliptical or punch, showing histological evidence for recurrence (with or without residual lesions), and contained the scars of the previous shave procedures. The excision slides were recruited along with the slides of the original SRS. The selected shaves were accompanied by Melan-A immunohistochemical stain (IHCS) to highlight the subtle tumor cells that might otherwise have not been identifiable by the routine Hematoxylin and Eosin (H&E) stain. The SRS in which lateral or deep margins could not be assessed due to orientation artifacts such as twisted specimens, absence of full face, or missing lateral edges, as well as the cases without Melan-A IHCS, were not considered.
All tissues were fixed in formaldehyde (10%) for 12–36 hours and were embedded in paraffin, and Melan-A IHCS (clone A103, 1 : 25, Dako, Carpinteria, CA, USA) was applied on every case after antigen retrieval. UltraView universal detection kit with multimer technology method was used on an autostainer (Ventana Benchmark XT, Tucson, AZ, USA). Then, the slides were retrospectively reviewed by one dermatopathologist (the author) to identify the recurrent and residual lesions on the ES and to determine the involvement of the margins of the original SRS. The measurements were all performed using a calibrated eyepiece containing a graduated horizontal scale.
The case group was defined as 110 ES that showed recurrence (with or without residual lesions), and the control group was defined as the patients who did not develop clinical recurrence after two years of follow-up (110 mildly atypical DN diagnosed on 92 patients). Grading of DN was based on architectural atypia and cytological atypia. Cytological atypia for mildly atypical DN was based on nuclear size (about the size of a keratinocyte nucleus), hyperchromasia, and the lack of conspicuous nucleoli, abundant cytoplasm, or pagetoid spread, which are the histological features of higher grade DN and melanomas [
In the ES (the case group), recurrence was defined as the presence of junctional melanocytic proliferation with or without upward migration, all limited within the boundaries of the underlying dermal scar, or the presence of dermal neoplastic melanocytes within the scar [
The presence of the neoplastic melanocytes in the adjacent unaltered epidermis or dermis was considered residual lesions. Dermal components showing benign nevus cells with histological maturation with descent located underneath the dermal scar were not considered as residual DN on ES but rather were considered as part of the background compound nevus.
On the original SRS, the true lateral borders of the lesions were defined as the last junctional melanocytic nest or as the last atypical melanocyte, either on H&E or on Melan-A IHCS. The true deep borders of the lesions were defined as the deepest extensions into the hair follicles or as the deepest dermal atypical or pigmented melanocytes (if there were any). Dermal histologically maturing melanocytes were not considered as the true deep borders of the lesions. The true lateral margins were defined as the edges of the specimens where both dermis and epidermis (dermoepidermal junction) were present (i.e., extension of the epidermis beyond the dermis was not considered as the true lateral margin).
On the SRS, negative (or clear) margins were defined as neoplastic melanocytes in the junction, as well as the cells that extended to the hair follicles or the deepest dermal atypical or pigmented melanocytes, confined within >0.2 mm of the lateral and deep specimen margins. Melan-A IHCS was performed on all cases, which highlighted the subtle junctional tumor cells that may otherwise have not been identifiable by the routine H&E stains. Positive margins were defined as extension or very close extension (≤0.2 mm) of the lesional cells to the above-mentioned structures or if epidermal rete ridges containing nevus cells were transected superficially (see Figures
Clear Margin. True lateral border of the lesion: the last junctional melanocytic nest or the last atypical melanocyte (arrow). True lateral margin: the edge of the specimen where dermoepidermal junction is present (arrowhead) (H&E; 20x).
Margin appears clear (negative) on H&E stain (H&E; 20x).
Melan-A highlighting subtle atypical melanocytes extending to the margin, not identifiable by the H&E stain in Figure
DN appears not to extend to the hair follicle (H&E; 20x).
Melan-A highlighting atypical melanocytes extending to the hair follicle at the deep margin, not identifiable by the H&E stain in Figure
In the second portion of this study, a retrospective review of 167 cases with biopsy-proven mildly atypical DN that were diagnosed between September and November 2013 and were followed up for two years was performed to determine the rate of recurrence.
This study was determined by Solutions IRB to be exempt from OHRP’s Regulations for the Protection of Human Subjects (45 CFR 46) under category 4.
Evaluation of the original SRS of 110 ES showing recurrence (the case group) revealed that 108 cases (98.2%) had positive margins as follows: 25 cases had extension both into the hair follicles and into the lateral margins, 69 cases showed extension only to the lateral margins, 9 cases showed extension only to the hair follicles, and 5 cases had none of the above features; however, the epidermal rete ridges containing nevus cells were transected superficially. No shave specimen showed dermal atypical (or pigmented) melanocytes extending to the deep margins. Twenty-seven of the ES showed residual atypical nevi in addition to recurrent lesions.
Evaluation of the 110 SRS from patients who did not develop clinical recurrence after two years of follow-up (the control group) showed that 28 cases (25%) had positive margins as follows: twenty-three cases showed extension only to the lateral margins, 3 cases had extension both into hair follicles and into the lateral margins, and 2 cases showed extension only into the hair follicles at the deep margin. No cases with superficially transected epidermal rete ridges containing nevus cells were identified, and no dermal atypical (or pigmented) melanocytes extending to the deep margins were noted (see Table
Excision specimens with recurrence compared with the original shave removal specimens.
Case group (recurred) | Control group (no recurrence or persistence) | |
---|---|---|
| | |
Lateral margins | 69 | 23 |
Hair follicles | 9 | 2 |
Lateral margins & hair follicles | 25 | 3 |
Epidermal rete ridges transected superficially | 5 | 0 |
Deep margins (pigmented or atypical cells) | 0 | 0 |
| | |
| | |
Of the 220 total shave specimens studied, in 68 cases (31%), Melan-A IHCS helped identify the true borders of the lesions which were not identifiable by the H&E stain alone.
Comparison between the original SRS and ES (see Table
Comparison between the features of the original SRS and ES.
Original SRS | Junctional ES | ES compound |
---|---|---|
Junctional, 39 | 38 | 1 |
Compound, 71 | 40 | 31 |
DN with positive margins defined by extension (≤0.2 mm) of the lesions to the lateral margins or into the deep margins through the hair follicles or when the epidermal rete ridges containing the lesional cells are transected superficially on the SRS have a statistically significant higher probability of recurrence than DN with negative (or clear) margins when followed up for up to 2 years (odds ratio (OR) = 158; 95% confidence interval (CI) = 36.62–683;
Follicular structures were present in only 81 shave specimens (37%), and 21 (26%) of them showed extension of the lesional cells into the deep margins. In the majority of those cases (15 cases, 71%), the neoplastic melanocytes were detected only by Melan-A IHCS. The author found Melan-A IHCS to be a useful marker to highlight the subtle neoplastic melanocytes at the dermoepidermal junction and their extension into the follicular structures, helping in identifying the true borders of the lesions that were not recognizable with the H&E stain.
As nevi (including dysplastic ones) mature in the dermis, they undergo senescence, defined by arrest in the proliferative capacity which is controlled by several mechanisms, such as telomere shortening, which is probably irreversible, and p16 expression. Senescence is accompanied by a number of morphological and functional changes, resulting in irreversible arrest in proliferation and pigment production of the nevus cells [
The data in this study differ from those of Sommer et al. [
Further, perhaps due to the higher numbers of recurrences and due to the utility of Melan-A IHCS, which detected more positive margins, the author found a more significant association between the margin involvement and recurrence compared to that found by Goodson et al. [
The author declares that there are no competing interests regarding the publication of this paper.
The author would like to thank Dr. Rami Geffner for his support in the preparation of this manuscript.